Forecasting Multiple Time Series with One-Sided Dynamic Principal Components

TL;DR

This paper introduces one-sided dynamic principal components (ODPC) for time series, enabling effective forecasting of high-dimensional data by using only present and past values, unlike previous methods that relied on future data.

Contribution

The paper proposes ODPC, a new method for forecasting high-dimensional time series, with proven convergence properties and superior performance compared to existing dynamic factor model approaches.

Findings

ODPC effectively forecasts high-dimensional time series.

Convergence of estimated values to population analogues is proven.

Forecasts with ODPC outperform other dynamic factor model methods.

Abstract

We define one-sided dynamic principal components (ODPC) for time series as linear combinations of the present and past values of the series that minimize the reconstruction mean squared error. Previous definitions of dynamic principal components depend on past and future values of the series. For this reason, they are not appropriate for forecasting purposes. On the contrary, it is shown that the ODPC introduced in this paper can be successfully used for forecasting high-dimensional multiple time series. An alternating least squares algorithm to compute the proposed ODPC is presented. We prove that for stationary and ergodic time series the estimated values converge to their population analogues. We also prove that asymptotically, when both the number of series and the sample size go to infinity, if the data follows a dynamic factor model, the reconstruction obtained with ODPC converges, in mean squared error, to the common part of the factor model. Monte Carlo results shows that forecasts obtained by the ODPC compare favourably with other forecasting methods based on dynamic factor models.